Stage

[Stage ] Real-time human joints kinematics estimation in sports applications using a stereo camera and deep learning

Équipes / Services concernés

Responsables

Mohamed Adjel / Kahina Chalabi / Vincent Bonnet

Date de publication

22.10.24

Prise de poste souhaitée

15.02.25

The general objective of this internship is to develop and validate a new framework for the real-time determination of human pose performing sport, rehabilitation and industrial tasks. The proposed method will be based on several deep learning trained models, a constrained biomechanical model of the human, and a(several) low-cost RGB-camera(s).

The internship will open opportunities to pursue a PhD in robotics/biomechanics within the GEPETTO team at LAAS-CNRS.

The details of the offer can be found here:

https://cloud.laas.fr/index.php/s/tnlg3O1Y6HvPudv